There are numerous benefits of computer-assisted interviewing (CAI) for
data quality and ease of interviewing, but one of the challenges CAI poses
is how best to represent the computer-programmed interview on paper.The
questionnaires for the NSFG Cycle 6 are available in
two formats:
CAPI-Lite format CAPI Reference Questionnaire (CRQ) format
Both formats represent the
basic content and routing of the full NSFG interviews, including the
computer-assisted personal interviews (CAPI) administered by interviewers
and the audio computer-assisted self-interviews (ACASI) that respondents
completed on their own. However, each format of the questionnaire
offers users a different level of detail and perspective on how the
interview was conducted.
CAPI-Lite format
The male and female interviews are shown in their
entirety, but with abridged representations of the question wording variants
and shorter descriptions of skip patterns through the interview. With this
format, users can still get a clear picture of how the questions were asked,
in what order, and of which respondents. The full male and female
interviews (male Sections A-K and female Sections A-J) are each available as
PDF files: C6female_capiliteMar03final.pdf C6male_capiliteMar03final.pdf
CAPI Reference Questionnaire (CRQ) format
The CRQ represents the fully detailed specifications
for the interview that NSFG staff provided to the computer programmers who
created the instrument using the Blaise software system.
All question wording variants are shown, along with
the conditions defining when each variant should be used.
“Flow Checks” specify the
precise routing through the interview based on earlier questionnaire items
so that the appropriate next questions for a particular respondent appear
onscreen. In addition, in some instances flow checks are devoted to the
creation of a new variable from one or more of the "raw," or "asked"
variables. These are called "computed variables" and are described in other
sections of the User's Guide. The flow check specifies in detail how these
computed variables were defined. A summary list of computed variables
defined in each questionnaire section can be found at the beginning of each
section's CRQ, and those that are “passed forward” to be used for routing
later in the interview are listed at the end of each section’s CRQ.
“Edit
checks,” programmed into the instrument,
attempt to catch and resolve data inconsistencies in the field, rather
than requiring resolution after data collection. These consistency checks
are generally located in the CRQ after the questions they are intended to
reconcile, and enabled the interviewer to return to specific questionnaire
items and correct them if necessary.
Use of additional survey aids, such as Show Cards,
Help Screens, and the Life History Calendar (female interview only), is
noted on individual questionnaire items. For example, if a
question-specific help screen (accessible via the F1 key) was available
for an item, the CRQ indicates “[HELP AVAILABLE].” If the item’s response
choices were to be shown on a Show Card in the interviewer’s show card
booklet, the CRQ indicates the number of the show card along with the
response categories.
The CRQ for each section of
the male and female interviews is available as a separate PDF file. (For
example, AfemC6CRQ.pdf is the CRQ for female Section A.)